¥¤¥ó¥¿¡¼¥Í¥Ã¥È·×Â¬¤È¥Ç¡¼¥¿²òÀÏ Âè9²ó

Size: px

Start display at page:

Download "¥¤¥ó¥¿¡¼¥Í¥Ã¥È·×Â¬¤È¥Ç¡¼¥¿²òÀÏ Âè9²ó"

ことこひのと
4 years ago
Views:

2 8 (6/8) : 2 2 / 49

3 9 : 3 / 49

4 ARPANET in / 49

5 4 ARPANET ARPANET in / 49

lumeta internet mapping http://www.lumeta.com http://www.

6 lumeta internet mapping / 49

7 IP 7 / 49

8 ( ) (L3) : : 7 Application Application 6 Presentation Presentation 5 Session Session 4 Transport Transport 3 Network Network Network 2 Data Link Data Link Data Link 1 Physical Physical Physical node relay node node OSI 7 8 / 49

9 (hierarchical routing) Autonomous System (AS): ( ) Keio University: AS38635 WIDE Project: AS2500 SINET: AS2907 AS AS 2 AS 3c 3a 3b AS3 1a 1c 1d 1b AS1 2a 2c 2b AS2 9 / 49

10 IGP (Interior Gateway Protocol): AS RIP (Routing Information Protocol) distance vector routing protocol (Bellman-Ford algorithm) OSPF (Open Shortest Path First) link state routing protocol (Dijkstra s algorithm) EGP (Exterior Gateway Protocol): AS BGP (Boader Gateway Protocol) path vector routing protocol 10 / 49

11 (topology) ( ) 2 IP 11 / 49

12 traceroute : skitter/ark (CAIDA): 20 iplane (U. Washington): PlanetLab DIMES (Tel Aviv U.) AS BGP : RouteViews (U. Oregon), RIPE RIS 12 / 49

13 traceroute IP TTL (time-to-live) TTL 1 TTL 0 ICMP TIME EXCEEDED IP IP TTL = 1 TTL = 2 TTL = 3 src ICMP Time Exceeded dst ICMP Time Exceeded ICMP Dst Port Unreachable 13 / 49

14 traceroute sample output % traceroute traceroute to ( ), 64 hops max, 40 byte packets ( ) ms ms ms 2 jc-gw0.iij.net ( ) ms ms ms 3 tky001ix07.iij.net ( ) ms ms ms 4 tky001bb00.iij.net ( ) ms ms ms 5 tky001ix04.iij.net ( ) ms ms ms ( ) ms ms ms 7 ge toknf-cr2.ix.singtel.com ( ) ms ms ms 8 p6-13.sngtp-cr2.ix.singtel.com ( ) ms ( ) ms ms ( ) ms ms ( ) ms ( ) ms ms ms ( ) ms ms ms 12 st1-6-bkk.csloxinfo.net ( ) ms ms ms 13 st1-6-bkk.csloxinfo.net ( ) ms ms ms ( ) ms ms ms 15 * * * 16 * * * 17 * * * 14 / 49

15 BGP AS AS AS AS : AS BGP : BGP : BGP table version is , local router ID is Status codes: s suppressed, d damped, h history, * valid, > best, i - internal, S Stale Origin codes: i - IGP, e - EGP,? - incomplete Network Next Hop Metric LocPrf Weight Path * / i *> i 15 / 49

16 RouteViews University of Oregon 10 : AS / 49

17 active BGP entries (RIB): 557k on 2015/6/ / 49

18 CAIDA s skitter/ark projects CAIDA skitter/ark: traceroute / 49

19 AS source: 2009 Internet Observatory Report (NANOG47) 19 / 49

20 AS source: 2009 Internet Observatory Report (NANOG47) 20 / 49

21 (node or vertex) (edge) : : ( ) (path): 2 (subgraph): (degree): ( ) ( ) Bellman-Ford Dijkstra ( ) ( : ) 21 / 49

22 Dijkstra 1. : = 0 = 2. : (1) (2) dijkstra algorithm 22 / 49

23 : Dijkstra $ cat topology.txt a - b 5 a - c 8 b - c 2 b - d 1 b - e 6 c - e 3 d - e 3 c - f 3 e - f 2 d - g 4 e - g 5 f - g 4 $ ruby dijkstra.rb -s a topology.txt a: (0) a b: (5) a b c: (7) a b c d: (6) a b d e: (9) a b d e f: (10) a b c f g: (10) a b d g $ a b 5 d c 3 4 e g f 4 23 / 49

24 : JR ( 100m) $ cat jr.txt # JR Yamanote line Osaki - Gotanda 09 Gotanda - Meguro 12 Meguro - Ebisu 15 Ebisu - Shibuya 16 Shibuya - Harajuku 12 Harajuku - Yoyogi Tamachi - Shinagawa 22 Shinagawa - Osaki 20 # JR Chuo line Tokyo - Kanda 13 Kanda - Ochanomizu 13 Ochanomizu - Suidobashi Sagaya - Yoyogi 10 Yoyogi - Shinjuku 07 # JR Sobu line Ochanomizu - Akihabara 09 Akihabara - Asakusabashi 11 $ $ ruby dijkstra.rb -s Shinagawa -d Ochanomizu jr.txt Ochanomizu: (94) Shinagawa Tamachi Hamamatsucho Shinbashi Yurakucho Tokyo Kanda Ochanomizu $ ruby dijkstra.rb -s Tokyo -d Shinjuku jr.txt Shinjuku: (103) Tokyo Kanda Ochanomizu Suidobashi Iidabashi Ichigaya Yotsuya Shinanomachi Sagaya Yoyogi $ 24 / 49

25 sample code (1/4) #!/usr/bin/env ruby # # dijkstra s algorithm based on the pseudo code in the wikipedia # # usage: dijkstra.rb -s source [-d destination] topologyfile require optparse source = nil # source of spanning-tree destination = nil # destination OptionParser.new { opt opt.on( -s VAL ) { v source = v} opt.on( -d VAL ) { v destination = v} opt.parse!(argv) } INFINITY = 0x7fffffff # constant to represent a large number 25 / 49

26 sample code (2/4) # read topology file and initialize nodes and edges # each line of topology file should be "node1 (- ->) node2 weight_val" nodes = Array.new # all nodes in graph edges = Hash.new # all edges in graph ARGF.each_line do line s, op, t, w = line.split next if line[0] ==?# w == nil unless op == "-" op == "->" raise ArgumentError, "edge_type should be either - or -> " weight = w.to_i nodes << s unless nodes.include?(s) # add s to nodes nodes << t unless nodes.include?(t) # add t to nodes # add this to edges edges[s] = {} # if edges[s] doesn t exit, initialize with empty hash edges[s][t] = weight # if this edge is undirected, add the reverse directed edge if (op == "-") edges[t] = {} edges[t][s] = weight # sanity check if source == nil raise ArgumentError, "specify source_node by -s source " unless nodes.include?(source) raise ArgumentError, "source_node(#{source}) is not in the graph" 26 / 49

27 sample code (3/4) # create and initialize 2 hashes: distance and previous dist = Hash.new # distance for destination prev = Hash.new # previous node in the best path nodes.each do i dist[i] = INFINITY # Unknown distance function from source to v prev[i] = -1 # Previous node in best path from source # run the dijkstra algorithm dist[source] = 0 # Distance from source to source while (nodes.length > 0) # u := vertex in Q with smallest dist[] u = nil nodes.each do v if (!u) (dist[v] < dist[u]) u = v if (dist[u] == INFINITY) break # all remaining vertices are inaccessible from source nodes = nodes - [u] # remove u from Q # update dist[] of u s neighbors edges[u].keys.each do v alt = dist[u] + edges[u][v] if (alt < dist[v]) dist[v] = alt prev[v] = u 27 / 49

28 sample code (4/4) # print the shortest-path spanning-tree dist.sort.each do v, d # if destination is specified, skip other destinations next if destination && destination!= v print "#{v}: " # destination node if d!= INFINITY print "(#{d}) " # distance # construct path from dest to source i = v path = "#{i}" while prev[i]!= -1 do path.insert(0, "#{prev[i]} ") # prep previous node i = prev[i] puts "#{path}" # print path from source to dest else puts "unreachable" 28 / 49

29 graphviz ( digraph finite_state_machine { rankdir=lr; size="8,5" node [shape = doublecircle]; LR_0 LR_3 LR_4 LR_8; node [shape = circle]; LR_0 -> LR_2 [ label = "SS(B)" ]; LR_0 -> LR_1 [ label = "SS(S)" ];... LR_8 -> LR_6 [ label = "S(b)" ]; LR_8 -> LR_5 [ label = "S(a)" ]; } 29 / 49

30 2: twitter : : Kwak twitter data 4000 : twitter degrees txt (135KB) 10,000 twitter degrees.zip (164MB, 550MB) 4000 numeric2screen.zip (365MB, 756MB) ID 1. twitter following/follower 10, following/follower CCDF X following/follower log-log ID 4. : 5. : : PDF SFC-SFS : / 49

31 twitter degrees.zip (164MB, 550MB) # id followings followers numeric2screen.zip (365MB, 756MB) # id screenname 12 jack 13 biz 14 noah 15 crystal 16 jeremy 17 tonystubblebine 18 Adam 20 ev 21 dom 22 rabble / 49

32 10,000 X following Y follower log-log 1 (0 ) CCDF X following/follower log-log 1 50 ID # rank id screenname followings followers aplusk TheEllenShow britneyspears cnnbrk Oprah twitter / 49

33 sort sort : $ sort [options] [FILE...] options ( ) -n : -r : -k POS1[,POS2] : (1 ) -t SEP : -m : -T DIR : : file 3 /usr/tmp $ sort -nr -k3,3 -T/usr/tmp file 33 / 49

34 : 1 # ruby autocorr.rb autocorr_5min_data.txt > autocorr.txt # head -10 autocorr_5min_data.txt T00: T00: T00: T00: T00: T00: T00: T00: T00: T00: # head -10 autocorr.txt / 49

35 : 1 # ruby autocorr.rb autocorr_5min_data.txt > autocorr.txt # head -10 autocorr_5min_data.txt T00: T00: T00: T00: T00: T00: T00: T00: T00: T00: # head -10 autocorr.txt / 49

36 k R(k) = 1 n n x i x i+k k = 0 R(k)/R(0) i=1 n 2n R(0) = 1 n i=1 x 2 i 36 / 49

37 # regular expression for matching 5-min timeseries re = /^(\d{4}-\d{2}-\d{2})t(\d{2}:\d{2})\s+(\d+)\s+(\d+)/ v = Array.new() # array for timeseries ARGF.each_line do line if re.match(line) v.push $3.to_f n = v.length # n: number of samples h = n / 2-1 # (half of n) - 1 r = Array.new(n/2) # array for auto correlation for k in 0.. h # for different timelag s = 0 for i in 0.. h s += v[i] * v[i + k] r[k] = Float(s) # normalize by dividing by r0 if r[0]!= 0.0 r0 = r[0] for k in 0.. h r[k] = r[k] / r0 printf "%d %.3f\n", k, r[k] 37 / 49

38 set xlabel "timelag k (minutes)" set ylabel "auto correlation" set xrange [-100:5140] set yrange [0:1] plot "autocorr.txt" using ($1*5):2 notitle with lines auto correlation timelag k (minutes) 38 / 49

39 2: : ifbps txt format: time IN(bits/sec) OUT(bits/sec) original format: unix time IN(bytes/sec) OUT(bytes/sec) OUT IN traffic (Mbps) /03 05/10 05/17 05/24 05/31 time IN OUT 39 / 49

40 2: mean stddev 350 Traffic (Mbps) time (2 hour interval) 40 / 49

41 2: # time in_bps out_bps re = /^\d{4}-\d{2}-(\d{2})t(\d{2}):\d{2}:\d{2}\s+\d+\.\d+\s+(\d+\.\d+)/ # arrays to hold values for every 2 hours sum = Array.new(12, 0.0) sqsum = Array.new(12, 0.0) num = Array.new(12, 0) ARGF.each_line do line if re.match(line) # matched hour = $2.to_i / 2 bps = $3.to_f sum[hour] += bps sqsum[hour] += bps**2 num[hour] += 1 printf "#hour\tn\tmean\t\tstddev\n" for hour in mean = sum[hour] / num[hour] var = sqsum[hour] / num[hour] - mean**2 stddev = Math.sqrt(var) printf "%02d\t%d\t%.1f\t%.1f\n", hour * 2, num[hour], mean, stddev 41 / 49

42 2: set xlabel "time (2 hour interval)" set xtic 2 set xrange [-1:23] set yrange [0:] set key top left set ylabel "Traffic (Mbps)" plot "hourly_out.txt" using 1:($3/ ) title mean with lines, \ "hourly_out.txt" using 1:($3/ ):($4/ ) title "stddev" with yerrorbars 42 / 49

43 2: Traffic (Mbps) Mon Tue Wed Thu Fri Sat Sun time (2 hour interval) 43 / 49

44 2: # time in_bps out_bps re = /^\d{4}-\d{2}-(\d{2})t(\d{2}):\d{2}:\d{2}\s+\d+\.\d+\s+(\d+\.\d+)/ # is Tuesday, add wdoffset to make wday start with Monday wdoffset = 0 # traffic[wday][hour] traffic = Array.new(7){ Array.new(12, 0.0) } num = Array.new(7){ Array.new(12, 0) } ARGF.each_line do line if re.match(line) # matched wday = ($1.to_i + wdoffset) % 7 hour = $2.to_i / 2 bps = $3.to_f traffic[wday][hour] += bps num[wday][hour] += 1 printf "#hour\tmon\ttue\twed\tthu\tfri\tsat\tsun\n" for hour in printf "%02d", hour * 2 for wday in printf " %.1f", traffic[wday][hour] / num[wday][hour] printf "\n" 44 / 49

45 2: set xlabel "time (2 hour interval)" set xtic 2 set xrange [-1:23] set yrange [0:] set key top left set ylabel "Traffic (Mbps)" plot "week_out.txt" using 1:($2/ ) title Mon with lines, \ "week_out.txt" using 1:($3/ ) title Tue with lines, \ "week_out.txt" using 1:($4/ ) title Wed with lines, \ "week_out.txt" using 1:($5/ ) title Thu with lines, \ "week_out.txt" using 1:($6/ ) title Fri with lines, \ "week_out.txt" using 1:($7/ ) title Sat with lines, \ "week_out.txt" using 1:($8/ ) title Sun with lines 45 / 49

46 2: Mon Tue Wed Thu Fri Sat Sun Mon Tue Wed Thu Fri Sat Sun / 49

47 2: n = 12 for wday in for wday2 in sum_x = sum_y = sum_xx = sum_yy = sum_xy = 0.0 for hour in x = traffic[wday][hour] / num[wday][hour] y = traffic[wday2][hour] / num[wday2][hour] sum_x += x sum_y += y sum_xx += x**2 sum_yy += y**2 sum_xy += x * y r = (sum_xy - sum_x * sum_y / n) / Math.sqrt((sum_xx - sum_x**2 / n) * (sum_yy - sum_y**2 / n)) printf "%.3f\t", r printf "\n" 47 / 49

48 9 : 48 / 49

49 10 (6/22) : 49 / 49

¥¤¥ó¥¿¡¼¥Í¥Ã¥È·×Â¬¤È¥Ç¡¼¥¿²òÀÏ Âè5²ó

¥¤¥ó¥¿¡¼¥Í¥Ã¥È·×Â¬¤È¥Ç¡¼¥¿²òÀÏ Âè5²ó 5 2011 6 8 : 1 2 / 31 : 3 / 31 ARPANET in 1969 4 / 31 4 ARPANET ARPANET in 1973 5 / 31 lumeta internet mapping http://www.lumeta.com http://www.cheswick.com/ches/map/ 6 / 31 IP 7 / 31 ( ) (L3) : : 7 Application